As mentioned above, one of the major bio-safety concerns regarding the field release of GE rice at the scale of commercial production is transgene flow from GE rice varieties to its wild and weedy relatives . The primary bio-safety concern is that transgenes introgressing into the weedy rice populations that grow intermixed with cultivated rice may lead to unwanted environmental or agronomic consequences . The extremely close genetic relationship of weedy rice to cultivated rice, their similar phenology, and their high cross-compatibility suggest that pollen-mediated gene flow from GE rice to weedy rice populations seems highly probable. This hypothesis is reinforced by the fact that the conspecific weed always occurs within or in near the edge of cultivated rice fields . Thus, many studies, descriptive and experimental, have been carried out to measure the extent of gene flow from cultivated rice to weedy rice, both by pollen and by seed. Results from many experimental studies demonstrate that crop-to-weed gene flow in rice occurs at a low per-generation rate . In general, the estimated frequency of pollen-mediated gene flow is low most likely due to the predominantly self-pollination feature and the short life of pollen grains of both cultivated and weedy rice. Low per-generation gene flow is expected to have significant evolutionary effects when it occurs year-after-year . Thus, weed drying rack it is not surprising that descriptive studies have revealed crop-to-weed gene flow is an important evolutionary force in shaping the genetic diversity and structure of weedy rice populations .
Pollen-mediated gene flow can occur in both directions, namely, crop-to-weed and weed-to-crop. But unless some of the harvested grain is replanted, the important maternal source of weedy hybrids will be the weeds. If GE rice is involved in the crop-weed gene flow and introgression, the resulted weedy hybrid progeny will contain one or more transgenes. Such recurrent gene flow and subsequent introgression will play roles in the evolutionary dynamics of the weedy populations. Commonly, antibiotic and herbicide tolerance genes are used as selectable markers in GE rice transformation. They also have been utilized as markers in various experiments to detect the frequencies of rice transgene flow . As shown in Table 1, molecular markers are also used to quantify of crop-to-weed and crop-to-wild gene flow . Gene flow results collected from these experimental studies are the first step for assessing the impact of cropto-weed and crop-to-wild transgene flow in rice. The earliest Oryza crop-to-weed gene flow field experiments are from Sanders et al. who used two herbicide-tolerant rice varieties in independent studies at the Louisiana State University Research Farm in the United States. In these experiments, weedy rice populations were interplanted with the cultivated rice varieties. The gene flow frequencies were estimated by screening seedlings from weedy rice parents with the appropriate herbicide; crop-weed hybrid seedlings would be expected to survive. Extremely few seedlings from weedy rice parents interplanted with the glufosinate-tolerant rice survived glufosinate screening. The few weedy rice plants surviving from the imidazolinone spray were found NOT to be products of natural hybridization. Thus, the researchers estimated that gene flow was extremely low .
Soon after, Estorninos et al. used microsatellite markers to determine the outcrossing rate between the imidazolinone-tolerant rice line “CL 2551” and awnless straw-hulled weedy rice at Stuttgart, Arkansas, USA; the outcrossing rate was estimated to be 0.0–0.05%. After reviewing more than ten published studies, Gealy et al. concluded that rice crop-to-weed gene flow was extremely low but highly variable. Later field experiments supported that conclusion. For example, gene flow frequencies from glufosinate-tolerant GE rice to nearby several weedy rice accessions ranged from 0.0 to 0.5% . Likewise, Shivraina et al. reported a comparable amount of crop-to-weed gene flow in a field experiment that allowed for measuring much greater spatial scale between source and sink plants; frequencies of 0.003– 0.008% crop-to-weed gene flow were recorded. Using a similar transgene marker, Jia et al. investigated the frequency of gene flow from the crop into two weedy rice populations: one in Leizhou of Guangdong province, the other in Yangzhong of Jiangsu province. The estimated gene flow ranged from 0.002% to 0.342%. Sun et al. assessed pollen-mediated gene flow from glufosinate-resistant transgenic hybrid rice to six weedy rice accessions, and found that the frequency of gene flow from GRrice to weedy rice accessions ranged from 0% to 0.47% in different designs of gene flow experimental. To date, the highest frequency of crop-to-weed gene flow was reported by Olguin et al. who studied the transgene flow from indica rice to 58 weedy rice accessions from Costa Rica; it was as high as 2.3%. Notably, Pu et al. also reported insect-mediated pollination in rice, where increased frequency of transgene flow was detected.
As a group, the above studies indicate that – although the gene flow frequency per generation is very low – it also varies considerably under different situations. Given that some gene flow is almost always present when the crop and the weed co-occur and that it will be recurrent if rice is planted in the same location every year, transgene flow and eventual introgression from a GE rice variety to weedy rice populations seems inevitable. It is probably impossible to stop the flow of GE rice transgenes into weedy rice populations without extraordinary efforts . But whether transgene introgression will lead to increased weed problems and/or environmental impacts also depends on any fitness or other evolutionarily significant phenotypic changes in weedy rice individuals that are associated with the introgressed transgene . If gene flow is expected to occur, then such data are next critical step in environmental risk assessment associated with transgene flow . Thus, the following questions need to be considered prior to commercialization of GE where weedy rice is present: What will be the agronomic and ecological consequences of introgressed transgenes in a weedy rice population. Will an introgressed transgene change the fitness and evolutionary dynamics of a weedy population?Measuring the fitness effect of a crop transgene in individuals with a history of crop-weedy/wild hybridization requires one or more field experiments that compare wild/weedy individuals in which the transgene is present and with those in which the transgene is absent. Such individuals can be created with artificial crosses of a transgenic crop line and its nontransgenic counterpart with their weedy relatives as illustrated in Fig. 2. Measurement of the comparative performance of the two types of hybrid progeny or their descendants in the field – such as the survival ratios, competitiveness, and fecundity under specific environmental conditions – reveals the fitness effect to the wild/weedy relative populations associated with the presence of the transgene . The fitness effects of a transgene are expected to be largely determined by the type of transgenes incorporated in wild/weedy population the type of anticipated selection pressure in the environment, rolling bench and the amount of admixture in the transgenic hybrid lineages . Therefore, three types of characteristics should be carefully used when estimating the potential short-term evolutionary dynamics and long-term ecological impacts of a transgene that has been incorporated in weedy/wild populations: possible positive or negative fitness changes associated with a particular transgene when any intended selective pressure is absent ; fitness changes under the intended selective pressure that is related to a particular transgenic trait in the intended specific environment; and fitness effects of a transgene in early and advanced generations of the transgenic hybrid lineage compared with their nontransgenic counterparts, including the pure wild/weedy ancestral types. This approach is in accord with the case-by-case principle for biosafety assessment of genetically engineered crops . We have conducted multiyear common-garden experiments to estimate the fitness effect of insect-resistance transgenes that transformed into cultivated rice and introgressed into weedy and wild rice populations to understand their potential impact. Regarding transgene flow to weedy rice, one of our major foci, we used a number of GE rice lines containing different insect-resistance transgenes to cross with weedy rice populations from different geographic locations. Bt and CpTI transgenic rice lines were developed to deter lepidopteran pests, such as rice stem borers and rice leaf-folders. Several generations of crop-weed hybrid progeny with transgene-present and transgeneabsent lineages were produced to study the long-term fitness effect of the insect-resistance transgenes. We compared an array of fitness-related traits for these hybrid lineages, as well as their crop and weedy progenitor under both natural-insect and low-insect pressure.With regard to herbicide-resistance transgenes, Oard et al. conducted a field experiment and evaluated the seed production, shattering, and dormancy in eight F2 populations produced from controlled crosses of two transgenic, glufosinate-resistant rice lines and four red rice biotypes.
The presence of the transgene was associated with significantly shorter plant height and different maturity in hybrids, compared to those in nontransgenic counterparts. In another study, Wang et al. carried out a multiple-year common garden study on the fitness effects of a transgene that confers tolerance to the herbicide glyphosate through overexpression of 5-enolpyruvoylshikimate-3- phosphate synthase . In a glyphosate-free environment, they found that the transgenic hybrid lineage had higher seed production, greater EPSPS protein levels, tryptophan concentrations, photosynthetic rates, and percent seed germination compared with nontransgenic controls. These results suggest that phenotypic changes associated with such an epsps transgene could result in faster establishment and increased competitive ability of the transgenebearing individuals compared to nontransgenic weedy rice and cultivated rice. Other field studies of weedy rice individuals with introgressed non-GE crop herbicide resistance have shown fitness advantages related with faster germination and taller plants . Gene exchange and the spread of herbicide-resistant alleles in weedy rice suggest that wide-scale adoption of transgenic herbicide-resistance rice as a means to control weedy rice will be limited unless biotechnological solutions could be used to significantly decrease the occurrence of gene flow or otherwise mitigate it . The next of generation GE rice in China is apt to involve multiple transgenes with different intended purposes traits to battle the complex of insect pests of rice and other yield constraints in rice . Such products of crop biotechnology will inevitably require even more complicated experimental approaches for environmental bio-safety assessment with regard to introgressed transgenes in weedy rice populations. More biotic and/or abiotic factors will have to be taken into account to simulate field conditions that may influence the fitness effect of the multiple transgenes employed simultaneously. The environmental bio-safety assessment of the consequences of crop transgene flow to weedy rice populations through artificially created crop-weed hybrid lineages and subsequent fitness testing has proven costly and time consuming , especially with regard to long-term evolutionary impacts. However, our multiple-year field experiment on insectresistant transgenic crop-weed hybrid progeny reveal that the experimental estimation of fitness effects should be sufficiently apparent based on data from hybrid lineage as early as two or three generations posthybridization . Also, it should be sufficient to make some preliminary conclusions on the impacts caused by the insectand herbicide-resistance transgene flow into weedy rice population case-by-case, based on results obtained from above studies. For the insect-resistance transgene, the fitness advantages brought by insect-resistance transgene might be limited due to the fact that weedy plants will be surrounded by insect-resistant plants in a GE rice field, and the extensive commercial cultivation of an insect-resistant GE crop will largely reduce target herbivores in a GE deployed area as previously described by Wu et al. for Bt cotton. As a result, insect-resistance transgenes flow into weedy rice will have a limited evolutionary impact. In contrast, herbicide-resistance transgenes should deserve more attention since the herbicide pressure in GE rice field will always favor the spread and fixation of any herbicide resistance transgene in weedy rice population, unless effective mitigation strategy to be developed to minimize the occurrence of gene flow or reduce the fitness of crop-weed hybrids. The control of weedy rice is challenging because of its unique characteristics, such as strong seed shattering and dormancy, abundant genetic diversity, and mimicry with cultivated rice varieties. Like other conspecific weeds, weedy rice can easily acquire genes/alleles from its cultivated progenitors through cropto-weed gene flow. Some crop alleles can also enhance the fitness of weedy rice, enabling it to adapt to and evolve rapidly in the cultivated rice agro-ecosystem. Although Crop-to-weed gene flow in rice occurs at a relatively low frequency , transgene introgression into weedy population is essentially inevitable because weedy rice and cultivated rice co-occur and year after-year in the same fields.